I’m : a programmer, writer, podcaster, geek, and coffee enthusiast.

Kindle Touch compared to Nook Simple Touch, Kobo Touch, and Kindle 4

Amazon’s new Kindle Touch, Barnes & Noble’s Nook Simple Touch (my review), and the Kobo Touch are surprisingly similar.

Nook Simple Touch, Kindle Touch, and Kobo Touch

They all have the same E-Ink Pearl screen with the same contrast, the same resolution, and the same type of IR touch-screen sensor. (Any screen differences in the photos are the result of uneven lighting, not any real differences.)

They’re all available for $99, but the Kindle and Kobo both show ads (“special offers”) at that price — if you’re looking for an ad-free reader, the Nook is the least expensive at $99, with the ad-free Kobo at $129 and the ad-free Kindle at $139.

Content libraries and ecosystems

All of the major e-readers have similar content libraries these days.

The Nook is particularly good for magazine availability, even slightly exceeding the Kindle’s availability in my searches. But Barnes & Noble’s content store is very buggy for me: I often get errors claiming unspecified problems with purchases, downloads, or connectivity. (This also happens on the Nook Tablet.)

Kobo’s ecosystem is still a disadvantage. In my searches, while book availability was pretty good, it had the highest prices most often. And critically, its magazine and newspaper selection is abysmal — if you intend to read magazines or newspapers on your e-reader, you shouldn’t consider Kobo.

Hardware and bezels

They’re all similar sizes. The Nook is the thickest and chunkiest. The Kobo is the lightest, at 185g versus 211g for the Nook and Kindle.

The thick IR-touch bezels are all deeper than on non-touch readers. The Nook’s is deepest and the Kobo’s is the most shallow:

From the top of the stack: Nook, Kindle, Kobo bezels under diagonal lighting

As I emphasized with the lighting angle here, the deep IR bezels cast shadows onto the screen margins. This is especially noticeable with close, single-point light, like a lamp on an end table or night stand.

The shadows amplify the perceived depth of the bezels, and they can be particularly problematic when they cover text or interface elements too close to the screen margins. The deeper bezels also make it more awkward to hit touch targets near the screen edges.

The Kobo’s touch sensor seems to work as well as the others, so I’m curious why the Nook and Kindle needed such deep bezels.

The non-touch Kindle 4

The 2011 $79 Kindle with buttons, which people have settled on calling the Kindle 4 (therefore, so will I), remains a strong competitor here, too.

Kindle Touch (left) and Kindle 4

Compared to all three touch readers, it’s cheaper ($79 with ads, $109 without), thinner, and noticeably lighter (166g).

It lacks the audio features of the Kindle Touch and previous Kindles, such as text-to-speech and MP3 playback, but I’ve never used those features or seen anyone else use them. (If you actually will use those features, the Kindle Touch is the only option for you.)

The Kindle 4 also has a shallower screen bezel than any of the touch readers:

From the top of the stack: Kindle Touch, Kindle 4 bezels under diagonal lighting

Again, this reduces shadows in the reading area and make it easier to read with light hitting the screen at an angle, which for me is almost always.

Page-turning and performance

The Kobo Touch is the slowest at page-turns, followed closely by the Kindle Touch. In fact, my old Kindle 2 turns pages slightly faster than the Kindle Touch. I’ll come back to that in a minute.

The Kindle 4 and Nook both turn pages at about the same speed (I can’t tell the difference, at least), and they’re much faster than the others. This is interesting: the Nook’s speed means that it is possible to make a touch-screen e-reader turn pages as quickly as one with buttons.

Of the touch readers, only the Nook has hardware page-turn buttons that you can optionally use, but they’re uncomfortable to use because they require too much pressure. On the Kindle and Kobo, you must turn pages by tapping the screen. I wish they all had good page-turn buttons — you’d think the hardware would be optimized for the most common action that people perform when using e-readers, but that’s unfortunately not the case.

With page-turn buttons, you can simply rest your finger on them while reading and push down slightly to advance to the next page. I miss this ability on the touch readers: you need to move your finger from wherever it’s resting (which can’t be the screen, of course) into the screen area and tap each time. My ideal e-reader would have a touch screen as responsive as the Nook’s and good hardware page-turn buttons, but that doesn’t exist today.

Don’t worry, fingerprints aren’t very noticeable on the touch readers. The matte e-ink screen surface minimizes their appearance, and they’re easily wiped off.

The mysteriously slow Kindle Touch

The entire Kindle Touch interface feels sluggish. Page turns, menus, and navigation all respond to touches only after lengthy delays. Its overall performance is similar to the Kindle 2.

Here’s a quick video of the Kindle Touch and Kindle 4, side by side, turning pages and then going to the home screen:

Page-flipping responsiveness of the Kindle Touch (left) and Kindle 4. Download.

E-ink is always sluggish compared to traditional LCDs, and the IR touch-screen could plausibly add latency to touch responses, but we know from the Nook that it’s possible to make a much faster touch-screen e-reader than the Kindle Touch.

It doesn’t seem like the touch input is at fault. Even pressing the home button on the Kindle Touch results in a long delay before the home screen appears. The Kindle 4 is much faster at that, and everything else. Presumably, the higher-end Touch doesn’t have a slower CPU, so this is likely a software problem.

The Nook Simple Touch wasn’t particularly fast when it launched nearly 6 months ago, but Barnes & Noble improved its performance with software updates, and it’s now just as responsive as the non-touch Kindle 4. I hope Amazon can achieve similar gains with a software fix for the Kindle Touch very soon, because its sluggish performance makes it difficult to recommend.

Text controls

Both Kindles’ text controls are mostly unchanged from the Kindle 3 (now renamed the “Kindle Keyboard”).

My impression of the Nook’s text controls still stands from the review: I can’t find a comfortable margin width or text size, since the increments available are too far apart. The margin is especially problematic: the smallest-margins setting, the only usable width in my opinion, puts the text too close to the deep bezel and its shadow.

The Kobo’s controls are very good, with very fine-grained increments for text size, margins, and line spacing. It’s also the only one to offer a toggle for text justification. Since it won’t hyphenate, its justification isn’t very good, but it’s nice to be able to turn justification off.

The Nook and Kindles often justify text with no option to disable it. The Kindles won’t hyphenate at all, and the Nook hyphenates too aggressively.

All of the readers except the Kobo use book-style indented paragraphs, while the Kobo uses web-style block paragraphs.

The Nook and Kindles offer the nice Caecilia font (the only Kindle font available before the Kindle 3), which I prefer to the other options available on those readers. The Kobo doesn’t, but its default Rockwell font is similar and is also highly readable on e-ink.

Touch versus non-touch controls

The biggest problems I keep running into with the touch readers are non-obvious tap zones and gestures.

When navigating lists and menus, I often find myself wondering where I’m supposed to tap, and whether I should tap, tap-and-hold, or swipe. This is exacerbated in magazines and newspapers, since reading them involves so much navigation.

The interfaces are more obvious on the non-touch Kindle 4, since hardware buttons and on-screen cursors and highlights make it more obvious what to do. But I also keep touching the screen and being disappointed that it “doesn’t work”.

I thought touch readers would be definitively better than non-touch models, but I was wrong: in reality, it’s a toss-up.

Magazines and newspapers

You’d think a touch e-reader would be a huge improvement for magazines and newspapers. Jumping between sections and articles can be faster on a touch screen, but it isn’t always.

Periodical navigation is most intuitive to me on the Kindle 4, followed by the Kindle Touch. I perform navigational errors more often on the Nook. (I didn’t test periodicals on the Kobo because the selection is so poor.)

This was a surprise: I expected the Kindle Touch to be the best for periodicals, but its poor performance hurts it, and navigating lists can be unintuitive.

Highlights, notes, and typing

The non-touch Kindle 4 is understandably awful at text input, requiring you to move a cursor around the on-screen keyboard with the directional buttons to select each letter (arranged alphabetically, not as a QWERTY keyboard), much like the painful process of entering text on a game console or an Apple TV. If you plan to do a lot of highlighting and note-taking, definitely don’t get the Kindle 4.

Highlighting and note-taking on a touch reader is much easier, but it’s still not as easy, responsive, or precise as on an iPhone or iPad.

Among the touch readers, the Kindle Touch has the worst text selection: you need to tap and hold on the first word, then drag to the last word and release. There’s no way to modify a selection if you want to expand or refine it, and it’s difficult to reliably get it right the first time. The Nook and Kobo both offer iOS-like “handles” on the sides of the selection that you can easily drag to refine it. The Nook is best here, beating the Kobo in responsiveness.

You really shouldn’t type on an e-ink device if you can help it. Like text selection, it’s nowhere near as fluid, accurate, and responsive as an iPhone’s virtual keyboard. Even a moderately fast iPhone typist will easily be able to outpace the e-ink refresh rate, leading to a lot of errors and delays to review what was typed.

If you must type a quick note, the Nook is again the best option, although it and the Kobo have an annoying grid keyboard layout. The Kindle Touch has a more traditional staggered-row layout, but its poor responsiveness interferes too much.

You could also consider the Kindle 3, which Amazon still sells as the “Kindle Keyboard”, since it still has a physical keyboard. Its tiny, clicky, grid-layout keyboard is pretty bad, though. The Kindle Touch’s on-screen keyboard might be faster for most people.

But all of these e-readers are very bad for text input. If you take notes while reading or otherwise intend to type a lot, an iPad running the Kindle, Nook, or Kobo app is probably a better choice than any e-ink reader.

3G and web browsing

All of these readers can connect via Wi-Fi for content downloads and web browsing (except the Nook, which lacks a browser).

Only the Kindle Touch and Kindle Keyboard offer optional 3G connectivity for an extra $50. It’s an unusual setup: there’s no monthly fee for data access, ever, for the life of the device. It just has connectivity as long as you’re in an area covered by AT&T. (It sometimes works internationally, but small fees are usually involved.)

On previous Kindles, you could use the 3G connection with the built-in web browser to have an effectively “free” cellular web browser indefinitely, but that party has mostly ended: the Kindle Touch can only use the connection to shop in the Kindle Store, receive new issues of periodicals (or Instapaper), or browse Wikipedia. Any other web browsing requires Wi-Fi. Only the Kindle Keyboard can still browse to any site over 3G.

This is mostly moot, since the web browsers are terrible. They’re all extremely slow and painful to navigate. Web navigation on the non-touch Kindle 4 and Kindle Keyboard are especially bad, since you need to move around a little mouse pointer with the directional buttons to click on anything, but I wouldn’t call the touch experience usable either.

E-ink web browsing is a novelty that you’ll probably try exactly once. But if you actually intend to do it a lot, and can’t or won’t buy an iPad, the Kindles have a better and more full-featured browser than the Kobo.

The value of the 3G Kindles, therefore, is primarily in their ability to download new periodicals and shop for new books away from home. If you travel a lot away from Wi-Fi, especially if you also subscribe to a daily newspaper or buy new books constantly, the 3G models might be worthwhile.

For most people, I don’t think the 3G is worthwhile. I thought I’d use it regularly on my Kindle 3, but in practice, I hardly ever did. I chose not to get 3G on my Kindle 4 and Kindle Touch, and I haven’t missed it yet.

Using Instapaper

If you’re trying to use my Instapaper service with your e-reader, the best choice is one of the Kindles. (Actually, the best Instapaper experience by far is on the iPad. But I’ll assume here that you’re interested in an e-ink reader.)

The only way to get Instapaper on the Nook or Kobo is to download an ePub file manually from the Instapaper website and transfer it over USB every time. This is tedious, and most people probably won’t make a habit of it.

If you have a Kindle (any Kindle), Instapaper can use Amazon’s document-delivery service to automatically send a new “issue” every day, every week, or on demand with a simple action on the website. This is much more convenient than manually transferring a file over USB whenever you want new content.

But Instapaper on Kindles isn’t perfect. I’ve been able to generate newspaper-style navigation that works on all non-touch Kindles, but the Kindle Touch and Kindle Fire use new periodical-navigation formats that I haven’t been able to successfully generate yet. I don’t know if Amazon will ever make this possible. And recent changes to Kindle “personal document” storage makes it difficult to delete old Instapaper articles from the Kindle Touch.

So, until (and unless) I can find ways around the Kindle Touch’s quirks, the best e-reader for Instapaper is the Kindle 4.


You can save $30-40 on the Kindle 4, Kindle Touch, or Kobo by opting for “special offers”, a euphemism for ads. I haven’t seen Kobo’s ads, but on the Kindles, the ads replace the pictures of classic authors on the sleep screen (see my Kindle 4 review), and they also appear as a bottom banner on the home screen.

On the Kindles, the ads aren’t intrusive, and if you get an ad model and change your mind later, you can just pay the difference and get the ads removed. So it’s not much of a risk to just get the ads.

If you’re getting it as a gift, though, you should probably get the ad-free models. Maybe it’s just me, but I’d feel tacky giving an ad-filled Kindle or Kobo as a gift.

If you’re going ad-free, the Nook becomes the cheapest: it’s always $99 and always has no ads.

What about the Kindle Fire, Nook Tablet, or iPad?

The $199 Kindle Fire and $249 Nook Tablet are also options. (So is the similar $199 Nook Color, but the Tablet is a much better product for $50 more.)

I reviewed the Kindle Fire and determined that it’s really not a good product, mostly due to very poor software and performance.

I haven’t written a review of the Nook Tablet yet, but after some light use, it seems like a much better product than the Kindle Fire. That said, it still shares many of the downsides if you primarily intend to use it for reading: these tablets are more expensive, bigger, thicker, heavier, and more complicated than e-ink readers, and they have less-readable LCD screens and much worse battery life. You can bring an e-ink reader on a week-long vacation and leave the charger at home, but tablets need to be charged almost as often as phones.

The iPad shares most of the other tablets’ downsides, but it does have the unique ability to run all of the e-reading apps: Kindle, Nook, and Kobo. (And iBooks, which only works on Apple’s devices, but doesn’t have noticeably different availability or pricing compared to Amazon or Barnes & Noble.) And it’s the best Instapaper reader.

A full-fledged tablet can be a better choice than an e-ink reader if you plan to watch videos or play games a lot, or if you need more general-purpose computer-like functionality such as email and web browsing. But these come at the expense of reading quality. For most people, the selection of videos and games is going to be much better on the iPad than on the Kindle Fire or Nook Tablet.

Despite the color screens and much better touch responsiveness, I can’t recommend the Kindle Fire or Nook Tablet for magazines or newspapers. The experience reading them is too poor due to sluggishness, bugs, poor navigation, and software complexity. The iPad is a much better tablet for periodicals.

If you’re primarily in the market for an e-reader, get an e-ink device. And if you’re looking for a tablet for apps, games, and videos in addition to reading, unless the iPad’s price is absolutely out of the question forever, I suggest getting the iPad.

So which e-ink reader?

All of these readers have exclusive features that I haven’t covered here, primarily because I haven’t needed to use them, and I bet most people won’t. In reality, what matters for e-readers most of the time is content availability, price, size, interaction (buttons or touch), and performance, roughly in that order.

The Kobo Touch is a decent product, and its hardware is quite good: it’s thinner and lighter, with a more shallow bezel, than the other touch readers. It’s almost as compact as the non-touch Kindle 4. But its advantages don’t outweigh its slow page-turns, often-higher book prices, and almost nonexistent periodical availability. So for now, I can’t recommend the Kobo.

The Nook Simple Touch’s fast performance and low ad-free pricing embarrasses the other touch readers, and its catalog is as good as Amazon’s. But it gets a lot of important details wrong, such as the poor granularity on text controls, the clunky swipe-to-unlock, and the buggy store. It’s also the thickest and widest, and its deep bezel casts the most shadows on the text. But it’s a good choice. If it was my only e-reader, and I didn’t want to use Instapaper, I’d be mostly satisfied.

The Kindle Touch is almost a very good product, but its poor responsiveness is distracting. The Nook shows that this probably could be fixed in a software update, but we’ll see if Amazon actually does: historically, Kindles have received very few software updates, and they usually don’t include major changes. The Kindle Touch is still a good choice, and it’s my favorite of the touch readers — but it just barely edges out the Nook. (And the Nook is six months old. I’m curious to see the next one.)

The low-end, non-touch Kindle 4 is actually my favorite e-reader today. It lacks the easier text selection and periodical navigation of the touch readers, and it’s effectively impossible to type on, but neither of those interfere with the most common actions when reading. It’s faster, thinner, and lighter than all of the touch readers, the interface makes the most sense and is the most responsive, and it works best with Instapaper.

If you’re ordering one of the Kindles from Amazon’s U.S. store, I’d appreciate if you’d use one of my affiliate links:

The others are available at their respective stores:

Special thanks to Kobo for supplying a no-strings-attached review unit.

iPhone Twitter app styles

Tweetie, later acquired and made the official Twitter app, has always been my preferred client. But with yesterday’s release of Twitter 4.0 for iPhone, most of Tweetie is gone, buried under cluttered styling and reorganized into marketing-speak tab labels.

One critical aspect of iPhone app style that usually gets too little attention is the side-margin width of the text. Tweetie always seemed the most platform-native and followed Apple’s conventions most closely, but Twitter 4 has dramatically widened the margins for no apparent reason, also necessitating a font-size reduction to fit (almost) as many words per line as before:

Twitter 4 (left), Twitter 3 (formerly Tweetie)

This is approximately in line with other good Twitter clients:

Twitterrific (left) with “Small” font, Tweetbot

But I’m really going to miss Tweetie’s wider text lines (along with many of its other great features, implementations, and design choices). On such a narrow screen, line width has a strong correlation to readability, and I think the other clients and Twitter 4.0 have gone in the wrong direction.

I put a lot of thought and experimentation into Instapaper’s default iPhone margin width, and I really think 10 pixels is ideal:

Instapaper 4.0 (left), Mail

The similarity to Apple’s own Mail app is probably not a coincidence. I look at Mail a lot — it’s the most-used app on my iPhone by far — and whenever I’m stuck and can’t decide how an interface or navigation element should work, I always look at Mail for guidance.

And in the past, I’ve always looked at Tweetie.

Tweetie was always a shining example — the best on iOS, I’d say — of making a very feature-rich, well-designed app that could (and did) win an Apple Design Award, but also didn’t go overboard with textures and reimplement too many standard UIKit controls and conventions.1 It proved that while good design often contains heavy graphical flourish, it doesn’t require it.

But now it’s gone. I’m sad not just for the loss of the second-most-used app on my iPhone, but also the style it exemplified — a style that I’ve always tried to create in my app, and by far the style that I prefer in other apps.

  1. This isn’t a dig at Tweetbot, but pointing out a general trend: iOS apps that win ADAs tend to use textures and custom controls much more heavily than Tweetie. ↩︎

The Cosmonaut stylus for touchscreens

Studio Neat, known for the Glif iPhone stand, recently finished their second product: the Cosmonaut wide-grip stylus for “touchscreens”. They’re trying to be inclusive, but I think it’s obvious that this is designed for the iPad, so I photographed mine in its natural habitat:

They explained the design on Kickstarter:

The Cosmonaut was born out of our desire to have a really great stylus for our iPads. We love to sketch out quick ideas or doodle on our tablets, and using a stylus is much better than a finger for such tasks. We bought several different models currently available on the market but they all suffered the same problem: they were designed to look and feel like a pen. But why? Writing or drawing on the iPad feels nothing like using a pen or pencil. For one, tablets are ideal for low fidelity sketching. Also, it is pretty awkward to rest your palm on the screen of the device because it throws off the capacitive detection. Writing on a tablet feels like writing on a dry erase board: fast, simple, low fidelity. The perfect tablet stylus is one that feels like a dry erase marker.

I backed it on Kickstarter as soon as I saw it last spring (along with a lot of other people), and it’s finally here. Anyone can order one, starting today, for $25.

I’ve tried a lot of iPhone and iPad styli, and I haven’t liked any of them before. But this one’s very different.

The capacitive screens on our favorite touch devices don’t respond to pressure, like old plastic-stylus PDAs. Instead, they can only detect a blob of electrically conductive material, like a human fingertip, being pushed against them.

The tips of most other capacitive-screen styli are little blocks of conductive foam. That arrangement sucks: it feels like writing with a sponge. And the little sponges are usually attached to too-short, too-lightweight barrels that just feel flimsy and unsatisfying.

The Cosmonaut is built completely differently, and it exudes quality.

There’s no foam anywhere in it. The soft rubber tip gives slightly when pushed, because there’s a small air pocket between it and the solid aluminum core. Here’s a cross-section photo:

Photo courtesy of Studio Neat, since I didn’t want to cut my Cosmonaut in half.
See more construction details in their video.

It doesn’t feel like a brick of solid plastic writing on glass, and it doesn’t feel like a flimsy sponge on a stick. It feels like a marker, as designed.

Writing with the Cosmonaut is satisfyingly firm and perfectly controlled, and it glides over the screen perfectly. I can’t really describe the gliding feel any better: it’s just perfect. When you use it for the first time, you’ll notice this immediately. (100% of the people in my household did.)

It’s noticeably heavy (45g), at about the same weight as five of these Pilot Precise V7 pens (49g):

The weight and thickness make it surprisingly easy and satisfying to manipulate. I have no trouble at all hitting tiny, precise touch targets very quickly even when I need to jump across the iPad’s screen to get there. But after a long doodle, I felt slightly fatigued, possibly because of the weight. (Someone who uses pens more than I do probably wouldn’t have this problem.)

Most styli with the traditional shape encourage you to hold them the same way you’d hold a pen: firmly and close to the tip. This often causes your hand to inadvertently rest on the screen, confusing the input to the app and often resulting in errors. The Cosmonaut’s size and shape encourage you to hold it the same way you’d hold a dry-erase marker when writing on a whiteboard: further up the barrel, with a loose grip, and with your hand floating far above the screen. I’ve never found myself accidentally resting my hand on the screen when using the Cosmonaut.

This is definitely for tablets only, as Studio Neat suggests. It technically works on the iPhone (and even my MacBook Pro’s touchpad), but it’s like using a big Sharpie on a Post-It note — there’s not enough space to do much, although you could use it in a pinch to operate an iPhone while wearing gloves. It works acceptably on the 7” Kindle Fire and Nook Tablet, but it really needs the space of the 10” iPad to shine for most uses.

So what are its uses?

I got to play with it for about ten minutes before my wife “borrowed” it. I didn’t see it again for an hour.

It’s just so damned fun.

The Cosmonaut is the only capacitive stylus I’ve wanted to use for more than thirty seconds. The others were novelties, but this is a truly useful tool. It’s excellent, and I’m going to start bringing it in my computer bag.

I’m not an artist, but I bet artists and doodlers will love it. It works amazingly well in Penultimate (pictured above).

I’m not a note-taker, but I bet note-takers will love it. It works very well with Note Taker HD, and MyScript Memo’s transcription is amazingly accurate.

I’m not a document annotator, but I bet… document annotators… will love it. (I don’t know. Someone’s buying those bazillion PDF-annotation apps.) I’ve already used mine to sign a PDF form in GoodReader.

And any Pictionary-like game would be a lot of fun with it.

But I think the best use of the Cosmonaut is exactly what Studio Neat intended: quick whiteboard-style sketching. That’s how I plan to use mine… once we’re done doodling in Penultimate.

Update, March 2012: The Cosmonaut is awesome with Draw Something on iPads.

Kindle sales numbers

Amazon issued a press release touting Kindle sales numbers yesterday: today announced that Kindle devices remain the hottest products this holiday season – for the third week in a row, customers are purchasing well over 1 million Kindle devices per week, and Kindle Fire remains the #1 bestselling, most gifted, and most wished for product across the millions of items available on since its introduction 11 weeks ago.

I have my own window into a tiny slice of Kindle sales, since Instapaper’s site has some moderately used Kindle-buying affiliate links, and my reviews of e-ink Kindles (which also contain affiliate links) get a lot of Google search traffic and seem to influence some people’s decisions. And my little window definitely doesn’t represent Amazon’s total sales:

Obviously, my numbers don’t line up with Amazon’s, since they claim that the Kindle Fire is the top seller. But I panned the Fire, was lukewarm on the Touch, and recommended the Kindle 4. And on Instapaper’s Kindle page, I labeled the Kindle 4 as the “recommended model”. So my proportions, especially how poorly the Fire is doing, only indicate that people using these affiliate links are listening to my recommendations, not that this is a remotely accurate representation of all Kindle buyers.

But I’m curious if the Amazon’s claim that the Fire is the top seller depends on any non-obvious technicalities, such as the distinction between ad-subsidized and full-price models within each family, or the distinction between 3G models. Every combination of ads, no ads, 3G, and no 3G shows up separately in the affiliate reports, and I have to imagine that they’re separate throughout all of Amazon’s internal stats.

I hope, for the sake of honesty, that Amazon’s claim about the Fire’s dominance is based on combined model-family numbers: that is, I expect from their press release, assuming the basic Kindle 4 with ads is the second-best-selling model, that the Fire is beating the combined Kindle 4 family (with and without ads), not just the one with ads.

Undocumented Nest incompatibility with single-stage wiring

Update: A new Honeywell has the same issue.

We’ve had two Nest thermostats for a few weeks now, and we mostly like them. But I’ve run into an issue with our wiring that Nest has all but confirmed, and I’ll need to either stop using the Nests or run new thermostat wiring.

Nest claims to be compatible with single-stage systems with two (heat only) or four (heat and air conditioning) wires:

When a thermostat calls for heating or air conditioning, it bridges the respective R supply wire to the call wire and the equipment turns on. Simple. (Thanks, Transwiki thermostat wiring page.)

This setup doesn’t provide any method for the thermostat to take power for itself from the system when the heat or air conditioning isn’t on: the only way is to draw some current off of the bridged circuits when they’re running, but they might not run often enough to keep a digital thermostat powered. That’s why most digital thermostats need replaceable batteries.

Another standard thermostat wire is often employed to solve this problem:

The Nest uses a built-in, permanent, rechargeable battery that automatically draws current from the system to charge itself. It claims to be compatible with systems that have this wiring arrangement even if they lack the C wire.

The problem arises when the Nest needs to charge itself and neither the heat nor air conditioning has turned on in a while, like on a mild day. Without a C circuit to take power from, it can only charge itself from running the system.

So it pulses the R-W heat circuit in short bursts to get power.

Maybe some systems are slow to respond to the call for heat, so this doesn’t result in anything noticeable. But my boiler reacts instantly, and it sounds like it’s really not enjoying this technique:

I’m not an expert, but I have to imagine that this is not good for my boiler’s longevity. And the noise it makes is annoying at best.

I emailed Nest support and they essentially confirmed that this is how it behaves, so it doesn’t appear to be a bug in mine. They recommended installing a C wire to solve the problem, which will require an electrician and, therefore, more money. I’m going to look into how much that will cost and decide what to do then.

But Nest shouldn’t claim compatibility with systems without a C wire. Their solution of pulsing the heat circuit clearly doesn’t work on all systems without significant negative side effects.

Update: A new Honeywell has the same issue.

Amazon Kindle Fire vs. iPad 2

Amazon published this Kindle Fire vs. iPad 2 page shortly after its announcement on September 28, 2011. I thought it was just a temporary promo, but it turns out that Amazon is still pushing it heavily.

It’s interesting that the quotes under “What People Are Saying” aren’t linked to their sources (Gizmodo, Extreme Tech, Ars Technica, Forbes/Mobiledia). Maybe it’s because all of them are simply reactions to the Fire’s specs and price after the announcement, not reviews from anyone who actually used one. By those standards, I could have been quoted along with them that day:

“The Fire will be the first Android-powered tablet to sell in meaningful volume… It’s definitely going to compete with the iPad” —

…but the context is far less effusive.

Why hasn’t Amazon updated the page with actual reviews since the Fire’s launch? Surely they can find some positive ones.

Anyway, the main attraction of Amazon’s Kindle Fire vs. iPad 2 page is the big comparison table. Other people’s feature-comparison checklists always leave out factors that are important to me, so I made my own additions that Amazon is welcome to include on their page:

Kindle Fire iPad
$300 less than iPad 2
Volume Buttons
Needs Improvement
Very Stable
Home Screen
That was a small swipe. Did you mean to tap? Try again
Tap an app to open it
Magazine Reading
Infuriatingly Awful
Pretty Good
No HD, no complete full-screen viewing, poor stability, and no convenient way to adjust the volume during playback
Very Good
HD-quality streams, stable app, volume adjustable via convenient buttons
Available Apps
Not A Lot
And mostly crap
A Lot
Critically acclaimed apps and award-winning games
Web Browsing Speed
Amazon Says It’s Faster
Reviewers disagree
Everyone Else Says It’s Faster
They must be fanboys
Overall Speed
Frustratingly Sluggish
Very Fast
Syncing Music And Videos From Your Computer
Only Manually
And 6 GB is really small
Automatically with iTunes
16–64 GB sizes available
Available Cases And Accessories
Very Few
3G Data
Not Available
Isn’t everyone always on a Wi-Fi network?
$129 extra plus $15/month for 3G service
Automatic Shutoff Feature
Sleeps automatically when idle or if you rest the bottom edge on a table or your pants or pretty much anything
Sleeps automatically when idle or when a supporting case is closed
Your Kids Can Charge Whatever They Want To Your Credit Card
And all purchases require your password
Only If You Hate Life
You Will Want To Throw It Out The Window
Kindle Fire iPad

Open versus closed headphones

Generally, open-backed (or “open”, or “open air”) headphones produce higher-quality sound at lower prices than closed-backed (“closed”) headphones, and all of the best-sounding high-end headphones are open.

But you probably shouldn’t buy open headphones. It’s irresponsible to discuss headphone recommendations without emphasizing the important distinction between open and closed, since most people aren’t familiar with how much this matters in practical use.

So I was disappointed in this explanation from Sam Grobart in the New York Times:

Open-backed headphones allow air to circulate, which is more comfortable. They also can help give the impression that sound is coming from around you, as opposed to emanating from your corpus callosum, which is a characteristic of closed-back and in-ear headphones. Closed-back models are better at sealing off the outside world, but your ears may get hot from the lack of air circulation.

That’s true, but “better at sealing off the outside world” needs to be expanded with a very important distinction.

Grobart’s right that open headphones, such as the entire Sennheiser 500-series and the great-sounding-but-uncomfortable Grado SR-60, do let a lot of outside noise in, so they’re terrible for blocking sound in loud environments.

But, critically, open headphones also let your music out. Even if you’re listening at a moderate volume, everyone in the room will hear a tinny, annoying version of your music. So open headphones are a poor choice for environments with people nearby, such as open offices, home use with anyone else in the room, airplanes, buses, or trains. (Don’t be that guy on the subway.)

That rules out a lot of the situations in which people use headphones. Therefore, generally, I don’t recommend open headphones. My best-sounding pair of headphones, the Beyerdynamic DT-880, is also my least-used because I’m hardly ever in my office alone and they’re too big to walk around with.

It’s also unfortunate that Grobart’s article didn’t even attempt to suggest some specific models as starting points. The article’s takeaway seems to be “do your own research”, but that’s not very helpful to average consumers facing the barrage of similar-looking models spanning huge price ranges from the high-end headphone brands.

So here’s where I think you should start:

The “non-portable” models are too big to walk around with. You can, but you’ll look odd, and you might need to wrangle a long, heavy, coiled cord. They’re fine for airplanes, though, if you’re willing to devote the carry-on space.

There’s a lot of Sennheisers on the list for good reason. I’ve had headphones from AKG, Grado, and Beyerdynamic that all sounded just as good (and sometimes better), but I’ve always found Sennheiser’s models to be more comfortable, practical, and durable.

How to make coffee when your house’s electricity is out

Instructions for Normal People:

  1. Drive somewhere and buy a cup of coffee.


  1. Wait until the electricity is restored before making coffee.

Instructions for Impatient, Geeky, Coffee Snobs:

  1. Light the gas stove with a match.
  2. Boil water in the Helvetica Kettle.
  3. Plug the coffee grinder into the APC UPS that still has some power left, turn it on, grind the coffee, then turn it off to conserve its power.
  4. Realize you had the wrong grind size, dump those grounds, fix the grind setting, turn the UPS on again, and grind the coffee properly.
  5. Brew with AeroPress.

Guess which method I chose.





Everyone has their bullshit. You can simply decide whose you’re willing to tolerate.